project

One reason to remove nearly all of the fat from a stock is so that the stock can remain fairly clear with a subsequent reduction. E.g., yesterday I had 4 quarts of fairly strong chicken stock, removed the fat, and reduced the stock to 3 C. Traditional separation technology consists of a sequence of simple separators. So, suppose we have separators k = -n, ..., -1, 0, 1, 2, ..., n in a line in this order. Each separator has one input and two outputs. One output is for, say, the water and the other is for the fat. The separators are not perfect. So the water output may contain only slightly higher fraction of water than the input; similarly for the fat output. We feed the mixture of fat and water into the input of separator 0. We get our most pure fat from the fat output of separator -n and get our most pure water from the water output of separator n. Otherwise, the water output of separator k goes to the input of separator k + 1, and the fat output of separator k goes to the input of separator k - 1. With n sufficiently large, this approach works really well, even for some things that are really, really difficult to separate. Hint: There is a really large such separating system in a really big building in Oak Ridge, TN, and they are not separating fat and water. Of course, such a system of 2n + 1 separators is intended for operating continuously where the large total volume of the material in process is not a concern. But, for separating fat from stocks, I borrow the principle of this approach of 2n + 1 separators. I start with a large kitchen spoon and some of the common 2 C Pyrex glass measuring cups with a pour spout. First I use the spoon to get essentially all the fat from the stock into glass cup A. In this, I may well get four times as much stock (say, in volume) as fat, but I do get essentially all of the fat. If for some reason I want the very last traces of fat, then I blot the surface of the stock with towels. Paper towels -- discard. Cloth towels -- toss in washing machine. This first step is progress because now all the fat is in cup A with maybe 1 C of stock, and the main volume of stock, possibly several quarts, is free of fat. For the second step, I pour the fat from the top of cup A into cup B. I pour slowly and carefully to move essentially only fat. When nearly all the fat has moved, I use a soup spoon to scrape the surface of the liquid in tilted cup A to encourage the last of the fat to move to cup B. It is possible to continue following the general approach of 2n + 1 separators above, but in kitchen practice at this point cup A is essentially free of fat and can have its contents added back to the main supply of stock. Also usually cup B has so little stock that I don't mind discarding it.

For a wok, I got a 14" diameter round bottomed hand hammered steel wok, made in China, from http://www.wokshop.com/ The 14" is the diameter at the top of the wok, that is, the largest diameter. They have still larger woks, but from taking a few measurements around my kitchen I concluded that 14" was about the largest diameter that would be easy to handle, store, rinse in the sink, etc. The material is apparently sheet steel, possibly the mild cold rolled steel used in auto bodies. So, no, it is not "cast iron". Differences: Cast iron has a relatively porous surface, and sheet steel has a much smoother surface. Cast iron is brittle, and sheet steel is not. The wok was likely formed by stamping, like auto body parts, but it may have been formed by spinning. I do not recall seeing any spinning lines. For the hand hammering, some small dents are visible. The dents don't hurt; I have no idea if they help. The handle is also steel. So, the handle looks like they took a piece of steel tubing, say, like used for auto exhaust pipes (for a small car), hammered one end to a flat tongue, and attached the tongue to the wok with three rivets. The rivets look like an embarrassing first attempt by a junior high student in metal working shop, but so far they have been plenty strong. For the free end of the handle, it looks like they used a tool to shrink the end so that the cut edges curled inside and out of the way. So, the end is smooth, not sharp. So far, it's great. Ordinary usage quickly got it with a nice black coating inside and out; so, it quickly got 'seasoned', essentially just from burned oil. I have done some Chinese cooking in the wok, but currently my favorite usage of it is browning the skin on pieces of chicken. I want to emphasize: For its intended purpose, it's great. It's rugged. So far the steel has not warped at all, even with high heat (see below). The surface is easy to clean, and food rarely sticks. The wok doesn't look like a symbol of a perfect wife's kitchen perfection in the mind of a bride dreaming of being the next Donna Reed, but for cooking, it's terrific. For aluminum or stainless steel, an enamel coating, a Teflon coating, etc., no way. I have an old aluminum wok with Teflon coating on the inside and enamel on the outside, and it was a big mistake -- a big waste. For a tool for handling the food in the wok, at a restaurant supply house I got a long handled all-stainless cooking spoon. I like this spoon, and so far I have not used the Chinese tools that I got with the wok. When cooking, I have an insulating cooking mitt on my left hand for holding the wok handle and have the spoon in my right hand. For cleaning, one approach is to put about 1 gallon of water in the wok, bring it to a boil, and use a long spoon to 'deglaze' the interior. The boiling water, then, goes 'over the side' onto the grass below. Surprisingly, the water has very little effect on the grass! But, I use the wok only about twice a month! A good alternative is to turn off the fire and just let the wok sit and cool while eating the food. When cool, just pour the water down the drain in the sink. But you won't find me carrying a gallon or so of boiling water in that wok to the kitchen sink! More careful cleaning is easy in the kitchen sink: I have a stainless steel scrubber that works well. It is really easy to get the interior back to a smooth flat black. For the exterior, I get off the worst of the spilled oil and otherwise leave the rest for the fire on the next usage. To stop rust, I do remove loose water; that precaution together with good seasoning and/or a residual film of cooking oil are effective, even on the handle which so far is not 'seasoned'. I store the wok in a cupboard and let it rest on the steamer insert so that the possibly somewhat still sticky exterior does not touch the cupboard surface. For a burner, some years ago I got a King Kooker propane burner, Model No. 88 PKP, from Sam's Club. Apparently King Kooker is still at http://www.kingkooker.com/ The box my unit came in claims that the burner puts out 170,000 BTUs per hour. Some arithmetic shows that 170,000 BTUs per hour would be just under 50,000 Watts. In comparison, the larger burner on a electric kitchen stove may put out only 18,000 BTUs per hour. The burner power level does adjust nicely and continuously. The lowest propane flow rates give a quite gentle flame. This burner was apparently intended for heating pots of shellfish in Louisiana beach parties. The construction is simple and rugged. So there is a simple frame of steel rods bent and welded. The burner is cast iron. A few of the minor burner pieces are stamped steel. The propane fittings are mostly brass. The Kooker does support the wok, but the support is not very stable. Thus, there are likely risks of the wok tipping and spilling its contents. So far I have not had that accident, but I would not want to use the wok for deep frying. A wok 'ring' might help stability. Of course, I only use the Kooker outdoors, on a high porch, fully open on three sides. My Kooker has been outdoors fully exposed to the weather for several years now with little apparent harm. Recommendation: If you leave such things outdoors, then be sure to cover the regulator valve and the propane tank valve so that rain, snow, etc. do not get into the valves, corrode, freeze, etc. I use aluminum foil to cover these valves and eventually discovered that my local squirrels are convinced that any piece of foil must contain tasty food and, thus, tear into the foil and sometimes carry off chunks. I don't think these squirrels learned this lesson in my neighborhood; maybe the squirrels migrate! I don't know what King Kooker is selling now. I understand that they have made some changes since I bought mine. Just before Christmas I looked at a propane burner being sold for deep frying turkeys outdoors. The burner looked quite similar to the one in my King Kooker. So, such a 'turkey cooker' might be a substitute. Still, I wouldn't try to deep fry over a propane burner, not even outdoors. Not me: My guess is that it would be all too easy to have some gallons of cooking oil at 400 F spill and get ignited.

Would like to underline Jackal10's outline of investing. In particular, mostly 'investing' is for high technology. The broad reason is that the examples of Intel, Microsoft, and Cisco suggest that the big returns are in high tech. That Wal-Mart, McDonalds, and FedEx did so well seems to be passed over. If you get the usual lists of venture capital firms, e.g., on Sand Hill Road in Palo Alto or on Winter Street in Waltham, their interests are essentially only in computing, software, digital communications, and biotechnology. For someone with a background in high technology, seeing exactly how restaurants do get funded would be interesting. Fat Guy gave some comments, but more details would be interesting. You might want to be careful about that $2.5 million: Even for high technology, that is a non-trivial amount of 'first round' 'equity funding'. Even if adding up all the remodeling, equipment, furnishings, etc. takes $2.5 million, maybe you don't need equity funding for all of that. E.g., you may be able to get commercial bank loans for some of the total, maybe vendor financing for some of it, your landlord for some of it, etc. You clearly know more about your business than I do and likely already thought of these things. On your side is, the US is awash in new restaurants, likely many more new restaurants each year than new companies funded by venture capital people. So, these restaurants get funded somehow! Apparently one major point in such things is publicity: So, it might help for you to get your name known where you are, on TV, in a book, etc. Your name can be your valuable 'brand', help bring in customers, and maybe help bring in investors.

Here're two better solutions: First, use ice cubes of ethyl alcohol! These freeze at -114 degrees C. When they melt, they take in as heat 24.9 calories per gram instead of 79.7 for water. For liquid ethyl alcohol, it takes 0.6 calories to raise one gram one degree C. So, to raise 1 gram of solid ethyl alcohol at -114 C to liquid at 10 C would take 99.3 calories which is almost exactly the 100 calories we got from one gram of water ice starting at -20 C. For the relevant target audience, the really big advantage here is that these ethyl cubes not only cool the drink well but, as they melt, make the drink stronger instead of weaker! Second, use dry ice and get a carbonated drink as a freebie! The temperature of dry ice (solid carbon dioxide) is -56.6 C. So, the easy way both to keep the dry ice and to make and keep the ethyl cubes is to get some liquid nitrogen at -196 C. Liquid nitrogen is made in huge quantities and is not very expensive. Uh, this is a joke, folks! Handling things that cold is dangerous! Leave such things to professionals, which may not include your local bartender! Also, 100% ethyl alcohol is strange stuff and likely dangerous to handle.

There is another problem: To cook the drink, we have to conduct heat from the drink to the 'ice' or whatever we use as a substitute for ordinary water ice. For this purpose, water ice is especially -- shockingly -- effective. If we start with 1 gram of water ice at a temperature of -20 C (temperature centigrade), then it takes about 10 calories of heat (small calories; the Calories in food energy are 1000 times larger -- sorry weight watchers!) to raise the ice to 0 C. Now something big happens: It takes about 80 (79.7 is more accurate) calories of heat to convert the ice at 0 C to liquid water at 0 C. Then it takes about 1 calorie to raise the temperature of liquid water at 0 C to water at 1 C, and similarly (approximately, it's quite accurate but not exact) all the way up to boiling at 100 C (assume ordinary pressure). Now, really big shock: It takes 539 calories to convert the water at 100 C to steam at 100 C. Then for steam (at atmospheric pressure) we are back to about 1/2 calorie per degree C. That's why it takes so much heat to reduce 4 quarts of stock to an intensely flavored delectable 3 cups! For granite, one Internet site claims that raising 1 gram of granite 1 degree C will take 0.19 calories. So, 1 gram of water going from -20 C to 10 C would use 100 calories, but 1 gram of granite going from -20 C to 10 C would use only 5.70 calories. So, per gram, water is about 100 / 5.70 = 17.5 times more effective than granite. So, instead of one pound of ice cubes, would need 17.5 pounds of granite cubes. Get a well built 'ice' bucket! Of course, granite is more dense than water; estimates from Internet sites include 2.4 to 2.8 times as dense as water. If we pick 2.6, then would need 6.7 times as much volume of granite. Also, make sure that 'ice' bucket for those granite cubes is bigger! Since the cube root of 6.7 is about 1.9, will need each dimension of the bucket to be nearly twice a long. If the refrigerator is not as cold as -20 C, then water ice wins by even larger ratios. Lesson: That 80 calories per gram for melting ice provides a huge amount of cooling per unit weight and per unit volume. Exercise: Consider cubes of lithium.

This list has some champion dieters. How do you do that? I don't have a clue. Years ago, I tried Atkins: Had a great time eating lots of Porterhouse steaks, broiled lobster tails with melted butter, omelets, etc. Spent a lot of money on food. Kept the grams of carbohydrates very low, plenty low enough to lose weight according to the Atkins book. Got the right color on the test strips. Continued for months but didn't lose any weight at all. Bummer. Years ago there was a canned product Metracal. There was a lot of milk in it. Bought cases of the stuff and nearly lived on it. Also ate Mrs. Paul's Fish Sticks (they were decently good then), lots of carrots, etc. Lost about 50 pounds. Then got weak. Once passed out, fell down, hit my head, got a cut. Went off the diet and all the weight came back at just 1 pound a week right back to where the weight had been and then basically just stayed right there. Perfect example of set point. Some years ago, there was a product from Carnation called Slender Bars. Then the ingredients were strong on peanuts, powdered milk, vegetable oil, and chocolate. One package of two bars was to be a single meal, and a box of four packages had a little less than 1000 calories. So, I ate a box a day plus a can of Campbell's Chunky Beef Soup and a 12 ounce can of V-8. I could lose about 25 pounds on that before getting too weak to function. Then Carnation changed the product and, I believe, eventually dropped it. Bummer. I tried running: I got up to 20 miles a week and was losing weight. Then my Achilles tendons developed calcium spurs. I had surgery to remove the spurs. Tendons stayed sore for years. Never could run on the tendons again. Bummer. I tried swimming. I had access to a nice pool, 100 meters long. Was able to get some decent exercise in that pool each night. Then the pool closed too early for my schedule. Bummer. I got a bicycle and started cycling. People threw bottles and cans; in cold or wet weather, not much fun; the bicycle rims kept getting bent on ordinary road bumps; the 10 speeds were too low which meant that even going up hill I needed a higher gear (I'm a relatively high torque low RPM kind of guy); from high torque, I broke the bicycle pedals; got stronger pedals; the bicycle seat cut off blood flow. Didn't lose any weight. Bummer. I got a high quality exercise bicycle -- solid accurate ergometer -- and got up to a reasonable level of exertion -- good resistance level, good speed, for a hour. Didn't lose any weight, and then noticed bicycle seat was cutting off blood flow. Gee, can cyclists be fathers? Changed seat; same problem. Have plans to construct a bench instead of a standard seat. Okay, here's a question: I don't have any problem eating less. Salads with diet dressing, lean meat from roasted chicken, fish cooked in microwave oven with soy sauce and garlic, apples, etc. are all easy enough to do. Have the big USDA nutrition book; now foods have good nutritional labeling; have scales, etc.; so it's easy enough to pick figures for total fat and total calories per day and meet them. Hungry? Yes, but I can put up with it, partly by guzzling 2 liter bottles of diet soda. But, it doesn't work. I get weak, too weak to function. Can't read or write. Can't think. Can't work. If push it, get horrible headaches that aspirin won't affect. If keep pushing it, get too weak to stand up; start fainting. In the end, getting too weak to function was the main problem in dieting. I have a friend that has lost a lot of weight. He keeps telling me how he does so well on fruit, chicken, salads, etc. I tried his things and got too weak to function within a few days. Bummer. In all the discussions about dieting, don't hear people talking about how they fight the problems of getting too weak to function, and to me that's the main problem, the only really difficult problem. Am I the only one that gets too weak, or do others have some ways around the problem?

sfmiller: "I find that the worcestershire/cayenne combination does a lot to cut against the inherent blandness of starch-based casseroles. I use it in macaroni and cheese all the time, and sometimes with scalloped potatoes. And don't underestimate the power of a handful of chopped flatleaf parsley (well, more than a handful, with your quantities), by itself or with another herb." WOW! While for my next batch want to consider all the suggestions that were given, was easily able to take a first-cut try at your suggestion of a Worcestershire and Cayenne combination! So, just scooped out about 1 pint, sprinkled on some ground Cayenne pepper and some ordinary (dark) Worcestershire sauce (don't have white W. sauce yet), mixed, heated it in the microwave, toasted some bread to help cut the excessive butter fat content in the last batch, and ate it. You're right! We're talking two swift kicks up some number of notches right away! It's nearly a different dish! All the outrageously strong onion, carrot, celery fragrance (and likely shouldn't forget the 1/2 C of minced garlic in the stock -- it really was a full 1/2 C, folks; I measured it; it was a somewhat packed 1/2 C; it was nearly all of two large heads of garlic) somehow go hand in hand with the W. sauce and C. pepper to carry the thing into something noticeable better! Some of the sauce near the edges my microwave burned a little was neat stuff! Ah, maybe I should get out a propane torch and brown the top! I also think that the W. sauce and C. pepper helped cut the butter fat flavor. The butter fat really should carry flavors and help make this a good dish; in principle, always reluctant to reduce the butter fat content! Look forward to getting some flat leaf parsley, fresh thyme, white W. sauce! The actual sauce making is fun: Get about 2 1/2 quarts of fragrant liquid velvet. It's going to be fun to stand there and add 1 t at a time B/W W. sauce and C. pepper, bump up the lemon juice and salt to match, dump in fresh herbs, mix, and pour over the rest. Maybe I'm beginning to see some flavor orchestration, things working together for much bigger effects! For "quantities", it's just one roasting chicken! I'm using a roasting chicken (actually, would like to use a stewing chicken) instead of a broiler or fryer in search of more flavor. I'm using the whole chicken and, thus, getting a lot of chicken broth effect. After I make a brown broth from the skin, bones, etc., there isn't a lot left to throw away! This is getting to be fun! eGullet -- good way to learn to cook!

Suzanne F: "Perhaps some chopped fresh herbs added to the final mixture? Tarragon if you like it, thyme if not. Chives." Fresh herbs, in the final dish? Wow! Didn't think of that. I just put the herbs in during the chicken poaching. Then I strained the stock through wire mesh and then cloth. So, all the herbs and pepper particles got strained out. My goal in the straining was to have just a well flavored liquid stock as a base for the sauce. From the browning and pepper, the resulting stock looked like coffee; it was dark but not cloudy. With all the white roux, milk, and cream, the final sauce did get an acceptable color. Usually I'm not too thrilled about tarragon, but my mind is not made up. Some of the fresh chives available look terrific (wrapped like long stem roses), and at times there are some good looking freeze dried ones. Some of the fresh thyme looks good, too. "Shiitake mushrooms?" Wondered about that. And oyster mushrooms. Not really going for a 'chicken-mushroom' casserole, but a wider variety of mushrooms might provide some of the needed extra flavors. "If adding Worcestershire, try the 'Worcestershire for Chicken' fka White Wine Worcestershire; sprightly but not as assertive as the regular kind." Haven't noticed that 'white' Worchestershire! I was guessing using the regular kind but not much of it. "Try turnip/rutagaba in its own before you add any. Very sharp flavor, as a member of the Cruciferae family. If adding, cut in small dice and only lightly blanch, to maintain texture. But if parsnips are too strong, so might turnips be." Super! I needed that! "Ham: if so, NOT smoked, or at least not highly smoked; it will throw off whatever balance you've achieved." Ah shucks! I was hoping that some smoke flavor might be just what I would need for some 'manly hunting lodge' effect! But smoke alone is simple, and really I suspected that it would be just crude and not interesting enough to make a good dish. Also, with the heavy browning that is giving the coffee colored stock, there is some browned/smoke flavor in there now, something like a low double bass note in an orchestra. sfmiller: "I find that the worcestershire/cayenne combination does a lot to cut against the inherent blandness of starch-based casseroles. I use it in macaroni and cheese all the time, and sometimes with scalloped potatoes. And don't underestimate the power of a handful of chopped flatleaf parsley (well, more than a handful, with your quantities), by itself or with another herb." WOW! You give me too much credit! I was just throwing out Worcestershire and Cayenne independently, not as an especially good combination! WOW, a special combination! Okay, I've now got two expert votes to get some fresh herbs in the final mix. Looks like I should do it! marie-louise: "I don't think noodle dishes reheat well; I also don't think they are good assembled too far ahead before baking (except for lasagna, that seems to work okay.) They're okay reheated, just never as good as when they are fresh. Somehow the noodles absorb too much of the sauce and the thing always seems kind of dry and bland. You lose that quality of tasting meat, noodles, vegetables and sauce as individual elements-it all blends together like a bland bowl of soup. If you really want to make this as a make-ahead dish, I'd prepare it to the point of adding the peas and noodles. Heat the chicken in the sauce, add fresh herbs and peas, then pour over freshly prepared noodles and bake (or serve as is.)" WOW! Such expertise! eGullet rocks! You're right: When I ate the excess that would not fit in my 8 quart pot, now that you mention it, there really was "that quality of tasting meat, noodles, vegetables and sauce as individual elements" and later "it blends together like a bland bowl of soup." That's what it did! And that blandness is much of what I'm trying to improve on. Gads, you know more about my dish than I do, and I've been eating it! WOW! And you are correct about the noodles absorbing (water from the) sauce: When I assembled the dish, the sauce, roux from 1 C of butter and 1 C of flour, 3 C of reduced stock, 4 C of milk, and 2 C of heavy cream, yielded a nice sauce, not goo, would coat the back of a spoon but was still clearly a sauce and not thicker. But, after reheating, it's all goo, no more liquid. I do blame the noodles: The trial I gave here was my first with noodles. In earlier trials, commonly on reheating the sauce thinned out. Indeed, my use of all that roux is partly in an effort to keep a nice coat the back of a spoon consistency on reheating; Pepin uses only a little roux, and when I followed his instructions fairly closely, on reheating the sauce had also thinned out. Looks like a summary remark is your "I don't think noodle dishes reheat well" WOW! I didn't suspect that! Suzanne F: "Marie-Louise, you're right. However, if you notice, project didn't quite cook the noodles all the way -- so they could finish cooking in the reheating. Although I find there's something oddly comforting about slightly mushy, slightly overcooked noodles. Especially if you don't bother reheating, and just eat the dish right out of the fridge." Yup, I've done that already -- eaten some right out of the fridge! Ah, Marie-Louise is still correct! The noodles don't turn to mush; the ones on top can even be a little crispy (desirable contrast for this dish); but the noodles pull enough water from the sauce to turn it to goo! So, Suzanne F, you are correct: The noodles did "finish cooking in the reheating" and dried out the sauce! marie-louise: "I always undercook any pasta that will be baked (or in this case, stirred stovetop.) I also make the sauce a little thinner, so once some of the liquid absorbs into the pasta, the sauce is the right consistency." I guessed and did the first but not the second! Again you know my dish without seeing it! The noodles do make the sauce too thick. So, if I keep the noodles in the refrigerated mix, then I should use less roux in the sauce. The quantity -- a bit over 5 pounds total -- of onions, carrots, celery, and mushrooms I poached in the broth to make a stock (reduced to only 3 C) made the sauce alone close to passing the KFC test, but including all of those vegetables in the final dish is a bit much. Gee, if I can't get this dish up from the bland category, then the only choice will be to top it with canned onion rings! Many thanks, all!

howard88: Gee, never used evaporate milk for anything! I wasn't really using the cream for viscosity, although it does give a special 'mouth feel'. With the 2 C of roux, there is enough viscosity. "turnip or rutabaga". Gee, I don't know anything about them. Should get some, do a Google on what is commonly done with them, taste them, and try them. Thanks. I omitted that once I tried parsnips -- they were too strong. Maybe I should try to borrow flavors from a traditional 'Coq au Vin'. Could fold in some brown glazed white 'boiler' onions, but they are a lot of work.

Did a Chicken Noodle Casserole. Below I outline what I did and ask for suggestions for the next step in making it better. GOAL: Goal is a pot of food for quick dinners: Keep big pot in refrigerator. For a meal, scoop out desired quantity, heat to about 170 F in microwave, and eat. Looking for something 'robust', say, hunting lodge food. BACKGROUND: Am borrowing from: Jacques Pepin, 'Jacques Pepin's Complete Techniques', ISBN 1-57912-165-9, Black Dog & Leventhal Publishers, New York, 2001. especially the dish "151. Chicken Pie" on pages 384 to 386. OUTLINE: Started with a Purdue Oven Stuffer Roasting chicken, 7.66 pounds as purchased. For liver, chopped, poached quickly in about 3 T of water, removed liver to a small flat dish, reduced poaching liquid and added it to the dish, let dish cool, and called kitty cat. Happy kitty cat. For the chicken, cut the wings into two pieces, cut the legs into two pieces, cut the breast from the back, cut the breast into two pieces. For the back, removed and discarded the 'red jelly stuff' under membranes on either side of the backbone near the tail. Also removed and discarded other loose material attached to the inside of the back. Heavily peppered chicken skin. Outdoors, in a Chinese steel wok, with about 1 C of Canola oil, over high heat from a propane burner with about 170,000 BTU/hour power, running at about 1/2 full power (low roar), quickly but darkly browned all the chicken skin. Also browned the neck, heart, and gizzard. Indoors, packed browned chicken pieces, along with cooking oil used, in an 8 quart pot containing also 1/2 C dried parsley, 4 dry bay leaves, 1 T dried thyme, 1 T dried rosemary, 1/2 C minced garlic, 4 C Chardonnay wine. Added water to cover. Got about 7 quarts total volume. Brought pot to simmer and simmered for 30 minutes. Removed meat and let drain in a colander set in a bowl. Poured stock through a strainer and discarded strainer contents. Removed and discarded fat from stock. To stock, added 2 pounds pieces of large yellow globe onion, 1 pound of carrot pieces, 1 pound of celery pieces, and sliced, washed large white mushrooms from a package with net weight 24 ounces. Total volume was about 7 quarts; there was just enough stock to cover the vegetables. Simmered for 30 minutes. Removed vegetables and let drain in a colander set in a bowl. Took apart chicken placing diced meat in one bowl and rest in another bowl. Got 3 pounds 1 ounce of diced meat. Breast meat was just cooked. Some of the thigh meat next to the bone was still red. Poured stock through a strainer; discarded strainer contents. Poured stock through cloth, twice; set cloth aside for washing later. Got about 4 quarts of dark fragrant chicken stock. Took stock from under colanders, removed fat, and added to main stock volume. Added to stock about 12 ounces of reduced brown chicken broth from earlier efforts. Reduced stock to 3 C. Result gelled at room temperature. Made white roux of 1 C unsalted butter and 1 C flour. Added simmering 3 C of stock all at once to hot roux and whipped until smooth. Added 4 C of hot whole milk and whipped until smooth. Added 2 C of heavy whipping cream and whipped until smooth. Added 1/2 C fresh lemon juice and 5 T salt. Sauce would coat the back of a spoon and tasted good. The onions, carrots, and celery gave a 'brightness' to the flavor. The mushrooms added a little contrast. Lemon and salt balanced each other and together brightened the flavors. There was too much lemon and salt for the sauce alone but hopefully about the right amount for the final dish. Browned and pepper flavors did not seem too prominent. In 1 1/2 quarts of simmering water, blanched 1 pound of fresh frozen petite peas. Discarded poaching liquid and added peas to vegetables. Earlier efforts showed that including blanching liquid from peas would add undesirable flavor. In 4 quarts of simmering water, simmered 1 pound of hard dry egg noodles for 6 minutes. Each noodle was about 1 1/2 inches long, about 3/16 inches wide, and twisted about 1 1/2 turns. Drained noodles in a colander; discarded liquid. Noodles were nearly fully cooked. In 8 quart pot, combined in layers chicken, vegetables, and noodles. Poured sauce over solids. Got a little too much volume for 8 quart pot -- ate extra. Covered pot and heated through, with occasional stirring, in 250 F oven to center temperature of 160 F. Chilled uncovered and then covered. Put skin, bones, etc., from chicken in pot, covered with water, simmered to make brown chicken broth to use next time. RESULTS: Eating quantity reheated in microwave, results: It's rich. All that butter and heavy cream are quite noticeable. There is a lot of bright flavor from the onions, carrots, and celery. The onions, carrots, and celery, in the traditional proportions of 2:1:1, balance each other so that it's tough to notice any one of the three. The peas are a nice contrast. Might use milk instead of the cream and make the dish more 'meaty' by reducing the proportion of vegetables and noodles. Otherwise, dish is a bit too 'vanilla', too simple. Dish could use some more flavor contrasts. Maybe the noodles might be dark green ones? One approach may be to get the wok smoking hot with only a little oil and toss the sliced mushrooms until edges brown. Earlier efforts showed that in a sufficiently hot wok, can actually get the mushrooms to brown before they can give up their liquid and that such mushrooms can add a smoky flavor to the dish. Maybe some different mushrooms would help? Sure, morel mushrooms, but my uncle hasn't sent any in a while, and the prices are a bit steep! Similarly for truffles! Maybe some bacon could be added somehow and would be good? Or, maybe there should be 1/2 C of diced ham? Would want some high quality ham with good flavor, maybe even with a maple and smoke flavor? Thoughts, suggestions for ham? Maybe stock should get 2 T of Worchestershire sauce? Maybe there should be 1 t of Cayenne pepper, 1 t of nutmeg, some sage? Dry red wine instead of Chardonnay? What might help? Thanks.

Busboy wrote: "It's been a long time since my econometrics and calculus days. Nonetheless, I think project's strategy may be unnecessarily complex because it goes beyond the relatively simple question "Do Jews as a group eat more Chinese food than similarly situated non-Jews?" to tackle the strength of the correlation between Jewishness (and other variables) have on propensity to consume Chinese food. Certainly this is a valuable figure to have if you are, say, trying to model the ideal density of chinese restaurants in a given area based on census data, but beyond the scope of the original inquiry." "Goes beyond"? not really. There are two aspects of Fat Guy's question that make getting an answer not so simple. First he asked "as a group". So, he's not asking if some Jews eat more Chinese food per person or if Jews in New York City do, etc., but what is the situation for all Jews? In principle he has to be including Jews in all countries, but I restricted to just the US and Canada. So, for an answer considering "as a group" somehow we have to consider all the Jews and all the non-Jews. E.g., maybe there is some place where Chinese and non-Jews really like the weather and eat a lot of Chinese food and some other place where there are a lot of Jews but the Chinese really hate the weather and there are no Chinese restaurants. Then our 'sampling' should include these people. The easy way around all such considerations is to get a list of all the people and select by throwing darts at the list. Since the dart throwing is independent of everything else, it all comes out in the wash (there are some theorems that say these things). Second, Fat Guy included "similarly situated". Here is an example of why: Maybe in the US and Canada, the Jews are mostly urban. And maybe everyone in urban areas eats huge quantities of Chinese food. In this case, being Jewish would have nothing to do with it. So, being Jewish would be 'spurious' and the real 'cause' would be living in urban areas. It may even be that Jews eat in delicatessens so much that in urban areas they actually eat less Chinese food than non-Jews but are so strongly urbanized that as a whole they eat more Chinese food than non-Jews. We could imagine similar effects from age, income, education, etc. So, Fat Guy is looking for one answer -- just one -- that applies to all groups of people, Jewish and non-Jewish, "similarly situated". With both "as a group" and "similarly situated", Fat Guy shows that he has had at least some contact with applied statistics essentially as in 'scientific' sociology. With these two aspects of the question, Fat Guy is coming close to trying to identify Jewish as a 'cause'. Here he is following the recommended path of the J. Coleman school of scientific sociology. Fat Guy proposed some analysis of variance with some special matching in the sampling. I proposed a 'regression' model to ease handling both variables with a few values and ones with many values. Also, I avoided the possibly onerous matching. In an important sense, analysis of variance is a special case of regression. In one place I did provide some detail: We can look at presentations of regression, e.g., C. Radhakrishna Rao, 'Linear Statistical Inference and Its Applications: Second Edition', ISBN 0-471-70823-2, John Wiley and Sons, New York, 1967. and see the assumptions listed. We can ask if our outlined use of regression would satisfy those assumptions? In particular, the errors, Z(i) in my description, are supposed to be independent, identically distributed, and have mean zero. I outlined how we could assume this. Using regression analysis without attention to the assumptions sounds like just throwing abstract nonsense at a practical problem; such things are done at times, and I wanted to avoid doing so. I was also more careful about a 'random variable' than is common. The elementary discussions of 'sample spaces' I have seen, typically for high school, actually get these foundations wrong; what they have won't work. I wanted to avoid such nonsense. I avoided mention of 'correlation' and 'propensity to consume' because they really are not very relevant and not needed, even in a fully detailed treatment. For calculus, it is not fully relevant: In some elementary treatments, calculus is used to define expectation for a random variable that has a density. In my treatment, I didn't have to assume that the random variables have a density. The way to define expectation in general is in the subject 'measure theory', and that is what Kolmogorov used as the foundations of probability. Measure theory is the grown up version of the core of calculus. For econometrics, that's by the economists, yet again borrowing with questionable appropriateness from areas of mathematics including regression, optimization, etc., and if all the economists were laid end to end ...! tryska wrote: "how about just walking into random chinese restaurants and asking the patrons if anyone of jewish descent would raise their hands. take note and then walk out." This approach would have problems with Fat Guy's "similarly situated". In parts of the social sciences, they pay attention to how quantities are measured. In particular they consider 'reliability' and 'validity'. Asking people just to raise their hands could yield questionable measures. Also, might want to count the number of people and the amount they spent. We should note that really we have not been very clear on 'amount of Chinese food'. One measure would be amount of money spent in US dollars, say, before tips and taxes. This measure does not promise to be exactly proportional to weight of the food, volume of the food, or total food energy in the food. Now, if we are going to get a grant for this research, then I believe that we will need new copies of SAS, SPSS, TeX, LaTeX, AMSTeX, Adobe's collection, Mathematica, and a few others, a new LAN with some laptops, desktops, wireless, servers, firewall, packet filtering, and an OC-192 line (would settle for 10 Gbps Ethernet). We will need a preliminary design, and for this we will need some 'participant observation' at a wide variety of restaurants, both Chinese and non-Chinese. The suggestions here for an international scope are well considered. We will need appropriate travel support and arrangements. Will need some gym time with personal trainers to work off all the 'sampling' efforts. Will need some good digital photography -- h'ware, s'ware -- to appropriately document our 'sampling'. For a start, I suggest a series of dinner discussions at 'Le Cirque'? We will also want to discuss 'exit' for this 'venture'. For the question, without "similarly situated", let's see: There are about 1 billion non-Jews that eat Chinese food essentially 100% of the time -- hmm ....

Experts in conducting surveys may have some clever ideas for how to collect the data.

Fat Guy: "Do Jews as a group eat more Chinese food than similarly situated non-Jews?" The responses you have gotten show a wide variety of opinions. One reason for 'science' is to have a relatively objective way to test different conjectures. Yes, for the parts of sociology -- e.g., the J. Coleman 'school' -- to be regarded as 'science', your question would be one of the first 'hypotheses' that would be 'tested'. We should not pass up the related question: "Do Jews as a group eat more Chinese food than non-Jews?" That is, leaving off "similarly situated" can yield a possibly important question. For an example, if we were allocating advertising resources for, say, soy sauce, should we seek to aim the advertising at Jews? That is, suppose we have two magazines, 'Good Housekeeping' and a similar magazine for mostly Jews. If we knew that Jews ate more Chinese food than non-Jews and knew little else, then we might aim the advertising at the magazine for Jews. To be more clear, we could agree that the whole 'market' was the US adult population; take a 'simple random sample'; for each person, determine Jew or not and amount of Chinese food eaten, and get two averages: (1) Amount of Chinese food for Jews and (2) similarly for non-Jews. Looking at these two numbers, we might then make our advertising decision. Here we would not much care about "similarly situated". So, why would we not do this? Or, why would we pursue "similarly situated"? Well, without "similarly situated", we would not be making much progress on 'causes'. One of the goals of scientific sociology was to identify causes and, then, move on to exploiting the causes to build predictive models. It is fair to say that here sociology had 'physics envy'. You mentioned one way to proceed: analysis of variance. For your question, however, a more likely path would be 'linear regression analysis'. There are close connections, but regression does better with variables that can take on many values. So, to proceed, we should identify some variables that characterize your "situated". The usual suspects are age, education, and income. Since fat guys may eat more than others, we might also include weight. Since we can't tell much about weight without height, we should include that, too. I would suggest including the population of the city in which the person lives. We are in the context of 'probability'. Since 1933 with A. Kolmogorov's work, the more serious work in probability goes roughly as: We imagine that we are performing an experiment. A 'trial' is performing this experiment once. Some sets of trials are of interest to us, and these we call 'events'. Given an event A, we believe that it has a 'probability' which we denote by P(A), a number between 0 and 1. Our probability P acts like a definition of geometric area. During the experiment we take measurements. Each measurement is called a 'random variable'. Actually a random variable is a function: Its domain is the set of trials, and its range is the set of real numbers. We can move on to define independence for events and random variables, distributions and expectations of random variables, etc. One point: If our experiment is to flip a coin 5 times, then that is just one trial, not 5, and the results are data on 5 random variables, not the values of one random variable on 5 trials. Indeed, with Kolmogorov's work, all we need see from all we do in this universe is just one trial of one really big experiment. For the coin flipping, typically we would assume that the five random variables were independent and had the same distribution. Again, suppose for our experiment we are interested in the population of the US and Canada. So, we can let random variable Y be the amount of Chinese food eaten (say, in the past year) by one person. We can let random variable X(1) be 1 if the person is Jewish and 0 otherwise. For some positive integer n, we can let random variables X(2), X(3), ..., X(n) be the values of the other variables we use, e.g., age, education, income, weight, height, and city population. Well, if each of Y and the X(j), j = 1, 2, ..., n, has an expectation, that expectation is finite, and the variance is finite (none of which have to hold in principle but all of which are quite reasonable in practice), then we can ask for constants a and b(j), j = 1, 2, ..., n so that Y = a + b(1)X(1) + b(2)X(2) + ... + b(n)X(n) + Z and where the expectation (mean value) of Z exists and is 0 and the variance of Z is as small as possible. Note: Sometimes constant a is written as b(0); the only difference is notation. So, with just these meager assumptions, the constants and Z will exist. The proof is short. The proof follows from a projection result in Hilbert space, and that result follows quickly from the parallelogram inequality. Projection is intuitive: If we pick a point in a room and want the point on the floor that is closest, then we drop a weighted string and see where it touches the floor. That point is the closest to our finger and is the projection onto the floor. If we pick a point on the floor in the corner, we now have three points and a right triangle and can apply the Pythagorean theorem, and the result is the usual 'sums of squares' quantities in the 'analysis of variance' table in regression. In Y = a + b(1)X(1) + b(2)X(2) + ... + b(n)X(n) + Z we have a 'linear model'. It may be that reality is quite different. That is, it may be that the X(j) really do determine Y but the equation is quite different from our linear one. Still, the constants for this linear model will exist as claimed. Thus, having these constants exist does not really mean that we have identified reality. In particular, neither can we conclude that we have found just the 'right' variables that 'drive' or determine Y. Instead, what we have said works for any variables X(j) at all (with the meager assumptions we listed). There is a sense in which a linear model is reasonable: Effects tend to be fairly 'smooth'. Commonly functional relations are differentiable, including in the stronger sense of Frechet in which case there is an excellent local linear approximation. Another view is: "I don't know about 'reality', and I don't much care. I just go searching. If I get a really good fit, then I figure I may have found something. Maybe I can take it to the bank. Or, sometimes linear models are the right ones. So, I will fit linear models. When there really is a good linear model, then I should get a good fit. So, a good fit is a necessary condition; it's not a sufficient condition, but it tosses out a lot that is not linear and is a good step toward the bank." Constant b(1) is what we want: It is the extra amount of Chinese food eaten due to a person being Jewish, imagining that everything else held constant. So, if we think of being Jewish being a 'cause' of eating Chinese food, then b(1) is the numerical measurement of this cause. Does this b(1) connect with the simpler measurement above, the two averages? Well, if X(1) is independent of each of X(2), X(3), ..., X(n), then yes, there is a connection (I omit the simple algebra). Otherwise, in general, no. In particular, it may be that changing from non-Jewish to Jewish and holding everything else constant cannot work in this problem; that is, in principle, the distributions may be that there are no such Jewish people. The mathematics is capturing this possibility correctly. For application to allocating advertising, b(1) may be less useful than the simpler analysis above. However, suppose X(2) is income. Some people might look at b(2) and see what one more dollar of income (everything else held constant) does to Chinese food eaten. If this number is large, then they might advertise soy sauce in 'Town and Country' or some such, especially if the woman on the cover that month is the daughter of a wealthy entrepreneur in China! In Y = a + b(1)X(1) + b(2)X(2) + ... + b(n)X(n) + Z can we regard b(j)X(j) as the 'unique' contribution to Y from our variable X(j)? If so, then we could drop some of our variables and the other b(j) would not change. If we were working with overtones in music, then that would be the situation because these overtones are 'orthogonal' and act like the three usual orthogonal axes in 3-space. Here, however, if we drop some of the variables, typically the remaining b(j) will change. When we are working with variables with discrete values, orthogonally can be fairly common. The above is all in terms of random variables directly. We have not mentioned data collection at all. Our results are a bit abstract but are important because they say that the relevant quantities do exist. One application is that we are now well on the way to showing that the most important of the usual assumptions in 'regression analysis' really will be satisfied. Sadly, this argument is rarely included in treatments of this subject. Mostly people are left wondering "How do I know I am satisfying the assumptions?" or take a macho approach "I will just push through, have the software do the arithmetic anyway, get something that 'fits', and to heck with the eggheads". We should note that we have assumed nothing about distributions except for means existing and finite and variance finite. We are not assuming that the expectation of Z is zero; instead we can prove that we can select the constants so that the expectation will be zero. We are not assuming that the distribution of Z is Gaussian; for the most important parts of the work, we do not need a Gaussian assumption. We might note that, while we said that the constants a and b(j) exist, we did not say that they were unique. To see that they need not be unique, just include some one variable twice; then we need only have the two new coefficients add to the one old one -- QED. In practice these constants typically are unique; sometimes in practice, especially with discrete data, they are not unique. This point alone is commonly misunderstood. That we have outlined how to show that the constants exist without assumptions needed to show that they are unique is elementary for people that work with Kolmogorov's foundations and Hilbert space and unusual otherwise. In particular, in principle we can get an equation that fits and predicts well but where the constants are not unique. This point is commonly missed in applied work. To continue, suppose for some positive integer m we survey m people. We assume that these people are independent and that the data we collect on them has the same distribution (we explain more clearly below). So, we survey people i = 1, 2, ..., m. On person i, we measure amount of Chinese food eaten Y(i). We regard Y(i) as a random variable; we will have a numerical value for it for person i in our trial. Similarly, we let X(i,j) be the random variable for variable j for person i. We are assuming that the joint distribution of Y(i), X(i,j), j = 1, 2, ..., n, is the same for all i. This assumption is fully reasonable. To see this intuitively, for the US, imagine selecting the people by throwing darts at a big list of US Social Security Numbers. Then we will have Y(i) = a + b(1)X(i,1) + b(2)X(i,2) + ... + b(n)X(i,n) + Z(i) with the expectation of Z(i) zero. Further, essentially because the people are 'independent', the random variables Y(i), X(i,1), X(i,2), ..., X(i,n), Z(i) will be independent of Y(k), X(k,1), X(k,2), ..., X(k,n), Z(k) The easy way to show this is from part of Kolmogorov's foundations. The intuitive view is important: What 'independence' means is, we are given the distributions and asked to guess the values Y(k), X(k,1), X(k,2), ..., X(k,n), Z(k) If knowing the values of Y(i), X(i,1), X(i,2), ..., X(i,n), Z(i) and being able to process those values in any tricky ways we want helps us, then there is some dependency; else we have independence. More specifically, let V(k) be the first list above of n + 2 random variables, and let V(i) be the second list. Then the list (V(k),V(i)) is independent. From the way we threw darts, independence is intuitively clear. Moreover, such intuitive arguments are the most important ones we have for establishing independence in practice. Independence is a powerful assumption and implies the weaker 'uncorrelated', and that is often enough for applications. To continue, we have argued that the constants a and b(j) exist. Good to know that there is a specific place we are going. Alas, likely in practice we cannot find the values of these constants exactly. But, with a large number of people m, we should be able to get some accurate estimates of these values. For the usual ways of getting estimates, we do meet the usual assumptions. In particular, the Z(i), i = 1, 2, ..., m are independent, have the same distribution, and have expectation zero. We know these things from what we have done. The derivations for the estimates are not difficult, but the notation is a bit much for this simple typing. Continuing, suppose we drop our data into some software (which does not have to be advanced, complicated, or expensive) and get our estimates. We will likely learn that our estimates are unique. Then we will look at the estimate we get for b(1). We can ask, if b(1) = 0 (that is, all other things being equal, being Jewish doesn't matter), then what is the probability of getting an estimate as large or larger than we did? So, we test the hypothesis that b(1) = 0. If the probability of getting such a big estimate is very low, then we reject that b(1) = 0. We can also get a confidence interval on our estimate of b(1). If we can assume that the distribution of Z(i) is Gaussian, then there are some classic approaches to these calculations. Typically it is reasonable to use these calculations just for 'ballpark' answers even if we are not so sure of a Gaussian distribution. There are also some 'non-parametric' approaches. If what really matters is city size, then our regression may show this. Hopefully the analysis outlined above would be able to test some of the conjectures given and improve on 'revelatory visceral reactions'. So, as for city size, in principle should be able to test Kosher, open on Sundays, fraction of income, amount of Moo Shi Pork, amount of Shrimp with Lobster Sauce, etc. Establishing 'causality' is usually challenging. I'm not sure I'd recognize it even if I saw it. The place I tend to conclude causality is when there is some physical or even mechanical explanation. So, for being Jewish being a 'cause', I might want to wait for something as physical as, say, DNA evidence. Using DNA to explain a taste for soy sauce promises to be a lot of work! Instead of some really strict versions of causality, it may make sense to consider weaker versions; some people believe this and cover blackboards with little directed connected acyclic graphs with nodes for variables and arrows for dependencies. Then maybe we could sort out Jewish versus Kosher, urban, open on Sundays, etc., that is, which one of these is 'fundamental' with the others in effect consequences of the fundamental one. If we are not interested in causality, then it is not so clear why we would be interested in b(1). I did regression for a few years before I studied Kolmogorov's work, etc., and the derivations I did for the above (not shown) were great fun. To me, then, the above is really clear and easy to understand. Some readers interested in cooking but not in mathematics and reading the above may understand the effect most recipes have on me!

Yes, there were no recipes, not even for Chinese food! Yes, the paper tries to be 'sociology'. I know little about sociology, maybe even less than about Chinese food, but my wife got her Ph.D. from essentially the most 'mathematical' and 'scientific' 'school' of sociology -- J. Coleman, etc. -- and I will never forget that case of "participant observation"! Since the paper appears to have been published, in perhaps a peer-reviewed journal of original research, I hesitate to comment. Fortunately, I was able to overcome this hesitation and continue! The paper starts out by quoting Max Weber. Yup, sounds like the authors are trying to suck up to the reviewers! Or, in the universal wisdom of one Julia Roberts movie, sucking up is what people really like! Notable is their: "APPENDIX: Research Methods Social scientists' discussions of methods usually focus exclusively on techniques used to identify and test hypotheses. We are including as well certain shared characteristics of our individual backgrounds and sociological orientations that helped turn our lifelong delight in Chinese food into a research problem that informed our approach. As Krieger suggested, we have found in conducting research in general, and especially in this project, that our visceral reactions to questions and data are often empirically relevant and theoretically revelatory." Yup, one of the scientific approaches and mathematical methods commonly proposed to be used in research in sociology is testing hypotheses. Yup, that paper has nothing to do with testing hypotheses. Their "exclusively" sounds like they want something else? Well, there is more mathematics! Ah, shucks! Wait 'till they get ahold of some of the other relevant mathematics! Poor people; I will feel sorry for them; I really will! Ah, "visceral reactions" are "theoretically revelatory"? Or, as a 'scientist' we should say, in the words of some fundamental Christians, or possibly Max Weber and his 'The Protestant Work Ethic and The Spirit of Capitalism', that they were not showing enough "work ethic", were being 'slackers', and were "back sliding"! Among some fundamental Christians, "back sliding" is something to be avoided and otherwise corrected ASAP! There are many examples of back sliding: Playing out of tune, eating too much, giving in to emotions, and the authors appear to be guilty of at least the last two! Shame! As science, the paper fills a much needed gap in the literature and would be illuminating if ignited! Apparently the authors took a wrong turn somewhere in the freshman registration line: They really wanted to pursue communication, interpretation of human experience, emotion with pathos and poignancy -- wanted to be English majors! The paper would have been enormously improved with a really good recipe and details for, and photographs of, General Tso's Chicken! Are the authors' "revelatory" "visceral reactions" the same as reading ones own entrails?

Noticed two more: resplendent overwhelming ah, for the table settings, the bouquets, the seasonal dining room decorations, etc. Sprinkle them around like powdered sugar! To get more 'reviews', just do a random permutation of the adjectives! Google can help: Just search on, say, 'redolent napery' and discover other similarly obscure pretentious verbiage. Ah, we could come out with a culinary vocabulary pretense ranking, test it for reliability and validity, etc.! Bruce is right; I was just having fun using too many adjectives! beans: "I enjoy your posts immensely and feel a bit tardy in saying so, but, WELCOME to eG!" Thanks. I was just afraid Steve or Jason would get torqued at my errant implacable incorrigible intransigent irascible irredeemable irrepressible irresponsible offensive outrageous recalcitrant truculent unconscionable uncontrollable unrepentant regarding sacred cows as the best hamburger! Ah, alphabetization automates alliteration! But eG is for whatever reason -- at least in part some excellent moderating -- a special place. The average verbal and literary abilities are awesomely high; some are much higher; I am way in the back of the pack. Also eGullet is attracting some serious expertise in food. I'm one that is grateful for the expertise and information that are here.

Pan: "I want to read your next review." Ah, I wore out my thesaurus on the last one!

Borris_A: "Machined interior? Is this something like forged pans?" No. You make a cast iron pan the usual old way of casting iron: Pour the iron into a mold of sand, perhaps clay (I'm not an expert on casting), break the mold, extract the casting. So, the surface of the casting has reacted with the mold and is different from the interior. So, often in casting, the surface is removed to leave just the cast iron underneath. In the US, long the standard cast iron frying pans were made by Griswold or Wagner. Apparently after they did the casting, they clamped the pan in front of a machine, essentially a milling machine with a cutter, and had the cutter go over the interior surface. So, the cutter made a pass over the bottom and the sides. The cutter left tiny circular lines but mostly the surface was quite smooth, smoother than what came from the mold. From all I have seen, now in the US nearly all the cast iron frying pans come from China and have the interior just from the mold and without the machining. So, the pans were cast and not forged. Anna N: "I say he's right on the button when he suggests that the modern home kitchen is largely a triumph of form over function." Or, the people selling items for a kitchen are totally convinced that they just must sell fantasy dreams and not practical tools. The ideal situation is a wealthy couple that has a gleaming kitchen, complete with a matched collection of hanging heavy copper pots, and dressed in formal evening clothes on their way out to dinner!

Enthralling experience, captivating romantic ambiance, shimmering scented fountains, redolent bouquets, ethereal evanescent aromas, quaint delicate bric-a-brac, antique scrimshaw, deeply upholstered seating nestled in discreet protected coves, rare old European wines, aromatic fine herbs, gleaming, sophisticated cutlery, elegant, sparkling crystal, exquisite, crisp, immaculate white damask napery, tantalizing appetizers, scrumptious entrees, andante intermezzi verdi, irresistible, luscious, wicked sinful tortes, tempting perfect fresh tiny wild raspberries adrift in clouds of cream, fragrant with whiffs of orange, punctuated with candied violets and crowned with threads of silver. Restaurant owners: Call about reasonable rates!

Fat Guy: Thanks for the estimate of 125,000 BTUs/hour for the power of the burners in US Chinese restaurants. That puts in perspective the King Kooker propane burner I have, claimed to have power of 170,000 BTU/hour. Uh, it's only for outdoors! If it's covered with snow, then the snow melts quite quickly once the burner is lit! This burner was made in Louisiana and is sold for cooking pots of shellfish at beach parties. It's crude but rugged and effective. The burners now being sold for deep frying turkeys seem to be similar. Over that burner, I heated my cast iron skillet too hot and cracked it. Going out to buy a new one, sadly discovered that these are now collectors' items and no longer sold new. Instead all we have are imitations that do not have a machined interior. So, at a local restaurant supply house, I got a steel saute pan such as you are describing, and, from a mail order vendor in San Francisco, I got a steel wok from China. The handle on the wok looks like something I would bang out of a piece of steel pipe in a few minutes in my workshop; the rivets look like a student's embarrassing first efforts in high school metal working shop. So far, the rivets are holding. Actually, the wok works well over the propane burner. My wok is now a fairly uniform smooth dark black on the inside. I've tried to imitate what the inexpensive US Chinese carryout restaurants do but have not found the right proportions of the ubiquitous chicken broth, soy sauce, vinegar, sugar, garlic, chili peppers, tangerine peel, black mushrooms, black beans, corn starch, etc. My stack of Chinese cookbooks tries to be too authentic and is not much help. Big bad bummer. Someday someone will actually write a book on what those restaurants really actually do and really actually show people how to do such things at home -- really, actually, not just some literary fictional fantasy. From what I have seen in, and cooked from, the books I have, one of those Chinese restaurants cooking such things would go out of business in a week. For the wok, I'm finding that it's terrific for browning chicken pieces. I put a lot of pepper on the pieces, put some oil in the wok, turn up the propane flow loud enough to hear, light the thing, assure my kitty cat that a fighter jet is not about to run into the house, wait for the oil to smoke, and brown the chicken pieces. The high heat does nice things to the pepper. Then I poach the chicken pieces, etc., and make a chicken casserole, last time, two chickens, about 7 quarts. I have yet to try the saute pan but may soon: Once again bought an aluminum saute pan with enamel on the outside and Teflon on the inside. Using that pan to fry hamburgers three at a time over the propane burner has nearly ruined the Teflon. So, should be moving to the saute pan soon. For the mysterious subject of 'seasoning', here is some simple advice: (1) When the pan is new, clean it once to remove any oil left from manufacturing. (2) Use the pan, for frying, with oil. (3) When cleaning the pan, just remove any liquid oil and stuck food and don't try to remove the burned oil. In the cleaning, it's okay to use some detergent, water, and abrasive, just don't try to get off all the browned or blackened oil. That's it: Keep this up, and soon will have a beautifully 'seasoned' pan. That is, for 'seasoning', basically just use the pan; that usage is enough to provide 'seasoning'. Seasoned steel is a terrific cooking surface. Over high heat, there may be a really big difference between a steel wok and many steel saute pans: The wok is beautifully convex on the outside with nearly constant 'curvature' everywhere. So, as heat makes the steel expand, the shape will remain convex. No way will 'dimples' appear. A new saute pan is also convex on the outside, but the curvature varies a lot, and usually the bottom is flat. So, as heat causes the steel to expand, no telling which way the bottom will bend. So, dimples are likely, and the result will no longer be convex on the outside. I still have one cast iron frying pan, just the right diameter for making crepes or pancakes. I cook the first side in this cast iron pan and cook the second side in a Teflon pan. I do use a Teflon saute pan for making omelets. For French toast, I have an old Westinghouse electric something that can cook four slices at once. The interior is a nearly indestructible rough non-stick surface. But I fully agree with you that a saute pan should be steel or cast iron; I want nothing to do with frying on a surface of aluminum, stainless steel, or enamel. For your main question "So why don't we see more of them in home kitchens?" Ah, that's just rhetorical, right? If not, then here's the answer. Not just everyone will see the answer right away, so I will have to explain! In the US there's an image, a norm, an expectation that the kitchen is the 'sphere of influence' of the 'little woman': She is supposed to be a good wife and mother; in the kitchen she is supposed to do well feeding her family; success here is nearly as serious as all of marriage, motherhood, and life; her pride, self-esteem, public image, and success in life itself are in the balance; here truly perfection is the minimum acceptable standard; cleanliness is next to godliness; and everything should be shiny, glossy, beautiful, and totally spotlessly clean. She should look like the image of ideal of motherhood, in long, soft, full skirts, that any child with a skinned knee would rush to. She should wear a perfectly white apron, the perfect whiteness a representation of the perfection of her duty and work. Then, in this context, a pan is a piece of sculpture to provide a communication, interpretation of those emotions, experiences. That's why for decades Farberware polished the outside to a mirror finish and used shapes and curves somewhere between Rococo and Donna Reed. And, now, you, Fat Guy, with "why don't we see", want to replace all those cherished images of marriage and motherhood -- the tiny delicate shoulders, pretty little hands, demure face, great sympathy, empathy, affection, and dedication, which must carry the future of our civilization -- with something burned, black, greasy and merely pragmatic that sounds like, and might have been stamped from, the hood of an old Ford. Shame, shame on you! Is nothing cherished anymore!

Dejah: The intended role of the geometry was not geometrically perfect dumpling pleats but merely a hopefully clear description of how to make pleats in dumplings. Alas, such a description could be clear only for people that really like geometry! Dumplings are items of enormous variety, and their shapes are not precise geometrically but quite amorphous. For my description, I learned from the help in this thread, from the photographs pointed to from this thread, another look at various dumplings I have bought, and some of what must be the situation from the basic geometry where an arc of about 220 degrees is matched and sealed with an arc of the remaining 140 degrees. The role of the pleats is to use up the extra 80 degrees. jo-mel: Yes, I would like to upload not just pictures but "samples"! Better check with Steve and Jason about software that permits uploading and downloading food samples -- just think of the incredible possibilities! Dejah: "Accept the challenge, PROJECT! please...." Gee, geometry is a powerful subject and one of my favorites (totally saved my tail feathers at times) but does have some limits to its powers! Again, for success with this challenge, will need cooperation from Steve and Jason -- "just a small matter of software"! Now, while I work on dumplings, here is an exercise for geometry lovers, and a solution will be at least as enjoyable as some steamed or boiled dumplings! Given a triangle ABC, construct a point D on side AB and a point E on side BC so that the lengths AD = DE = EC. When you do this, don't check with my geometry teacher: She never understood the solution! She said "You can't do that!", but she was wrong. There is a quote that the violinist J. Heifetz once described music critics as "The words without the music." Well, until Steve and Jason get their software upgraded, eGullet is the words (with some pictures) without the samples! Yes, we want the samples, but, borrowing from that movie about T. E. Lawrence in the desert, before the samples must come the words -- and maybe even some geometry! Thanks again for the help.

Dejah: No, pictures would be inconsistent: When instructions are given just in words, so should the corresponding results! So, for the geometric perfection, if any, of pleats, will have only words.

Ah, my description of how to fold and seal a dumpling in terms of geometry may not be for everyone! But to me, that is a description that is easy to understand! One goal I would have, especially just in my own notes, is to get the ideas described in words and without having to use diagrams, photographs, full motion video, or in-person lessons from Chinese grandmothers! In a broad sense, one of the most serious problems in wider dissemination of knowledge about cooking is the strong dependence on learning person to person instead of via written descriptions. Alas, descriptions for dumpling making are something like ones for putting on a coat -- much more easily observed than written! For pleats, I thought that they were standard! But, then, when I was in college I had a girlfriend who had a skirt with pleats and ..., but that's enough of that! Good to learn that I should not take too seriously making the wrappers thinner near the boundary of the circle. Thanks to everyone for the help. Will report back after I have some good results!

For the plane of the arc in the final dumpling, I'm concluding that it is vertical. So, when a dumpling is being served, the arc touches the plate only at its ends and otherwise is in the air (like an arc in architecture). Or, the arc is not fully in contact with the surface of the plate. Okay. I believe I am understanding the pleats. For the pictures, I downloaded all of those and looked at them at 100%, 200% and 300% magnification. I believe that the pictures needed better lighting and many more pixels, but from those pictures, I believe I am seeing how the pleats work. My explanation, from my current understanding, of the pleating is: We start with the dough: It has been rolled until it is a thin layer in the shape of a circle. The diameter of the circle is commonly about 3 inches. Some people recommend that the dough be thinner near the boundary (edge) of the circle; once we see how the pleating works to create extra layers of dough near the boundary, having the dough at the boundary start thinner would help avoid having the dough too thick. Also, there is some variation in how thick the dough is otherwise. Some 'delicate' dumplings have dough that is thin and nearly transparent. Other dumplings, more suitable for boiling, have dough that is thick enough be solid white and opaque. And I have had some examples where the dough was so thick it suggested an effort to consume extra flour! To describe in words and without drawing a picture, we should get oriented: We can think of a compass with directions north, south, east, and west with headings of 0 degrees due north, 90 degrees due east, 180 degrees due south, and 270 degrees due west. Suppose we are facing north and the dough circle is on a horizontal plane, our cutting board, flat in front of us. We find the center of the circle and draw the east-west line through this center. The line intersects the boundary of the circle on the west at point A and on the east at point B. From the center we move west along line AB about 1/2 inch to point C. We put the center of the filling over point C. So, we don't really put the filling in the center. We draw a north-south line through point C. This line intersects the boundary of the circle at compass heading of about 200 degrees at point D and intersects the boundary of the circle at compass heading of about 340 degrees at point E. Now, to close and seal the dumpling, we start by wetting the top surface of the dough near the boundary and along the arc DAE. When we seal the dumpling, we will move eastern most point B on the circle to western most point A. We will essentially fold the dough along the line DCE. So, in this sealing operation, we will have to join the arc DBE of about 220 degrees with the arc DAE of about 140 degrees. So, we have 80 degrees of extra arc to handle, and this 80 degrees of arc is what causes the pleats. Thus, it is the 220 degree arc DBE that gets pleated to shorten it by about 80 degrees so that it will match and seal with the 140 degree arc DAE. In this folding and sealing operation (here we are concentrating on the geometry; there is a more ergonomic description below), the part of the circle bounded by arc DAE and line DCE remains on the cutting board. The rest of the circle is lifted from the cutting board, essentially folding about 90 degrees along line DCE. Okay, suppose, then, we lift eastern most point B and position it over western most point A. We do not join (seal) these two points yet. Instead, we join the dough along the arc a few degrees counter clockwise (CCW) from point E and a few degrees clockwise (CW) from point D. So, points D and E will form 'points' or 'ends' of the final dumpling -- we seal at points D and E first. Then we are left with the 80 degrees of extra arc to fold into pleats. All we really need are just two pleats. One pleat starts a few degrees CCW of point E, and the other starts a few degrees CW from point D. At a pleat, the dough from arc DBE is used to make three layers of dough instead of just one. So, in a pleat there are two extra layers of dough. In two pleats we want to handle (use up) 80 degrees of arc. So, 80/4 = 20 which means that a pleat should be about 20 degrees of arc. So, we will seal for about 25 degrees CCW from point E and start a pleat and seal about 25 degrees CW from point D and start a pleat. So, starting at point E, here is how the dough on arc EBD goes: First the dough is sealed CCW for about 25 degrees; then the pleat starts and the dough is creased (folded 180 degrees) along a radius of the circle and goes CW for 20 degrees making the first extra layer; then the dough is creased (folded 180 degrees) along another radius and goes CCW for another 20 degrees making the second extra layer. Starting at point D, the work is similar: we seal for 25 degrees CW, crease 180 degrees along a radius and seal for 20 degrees CCW, crease 180 degrees along a radius and seal for 20 degrees CW. So, of the arc ADE of 140 degrees, we have sealed 50 degrees and have 90 degrees to go. So, we have an arc of 45 degrees either side of point A to join with an arc of 45 degrees either side of point B; we are joining two arcs, each of 90 degrees. So we need no more pleats and the arcs fit. For an 'ergonomic' method, we will hold the dough so that line DCE is horizontal and left-right in our hands, likely resting on our two third (ring) fingers, possibly aided by our fourth (little) fingers; point A is vertically above point C and being held there as the dough rests on our second fingers, and arc DBE is toward us. Our first (index) fingers are free. For the seals of 25 degrees CCW from point E and 25 degrees CCW from point D, we make those at the same time using thumb and index fingers. So, again, we start sealing not with points A and B but from the 'points' (will be points in the final dumpling) D and E. Then, to form the first extra layers of 20 degrees of arc each, for the 180 degree radial crease, we wrap the dough around our index fingers. Again, starting at point E, the dough of the long arc EBD goes CCW for 25 degrees behind our right index finger, for the first extra layer, makes a radial crease of 180 degrees by wrapping around the tip of our right index finger, goes for 20 degrees along our right index finger, and, for the second extra layer, makes its final 180 degree radial crease. So, the two extra layers are between the right thumb and right index index finger. With the two extra layers between the thumb and index finger, we can squeeze and join the second and third extra layers. Then we can move our index finger to behind the arc from E to A, squeeze with our thumb and index finger again, and complete the pleat. At the time, we do the same with left hand. Now all that is left is to join the remaining 90 degree arcs. It might be all wrong, but that's my understanding so far.

Counted 80 cookbooks, if we get to count books on wine. Now, you understand don't you see that, uh, I'm not proud of all of these books. Beyond just the number of books, what about the distribution of times of the day when people were posting their responses to this thread? Answer below. 01 AM 08|XXXXXXXXX 02 AM 02|XX 03 AM 04|XXXX 04 AM 07|XXXXXXXX 05 AM 10|XXXXXXXXXXX 06 AM 03|XXX 07 AM 33|XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX 08 AM 30|XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX 09 AM 43|XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX 10 AM 37|XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX 11 AM 54|XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX 12 AM 09|XXXXXXXXXX 01 PM 47|XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX 02 PM 37|XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX 03 PM 43|XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX 04 PM 55|XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX 05 PM 39|XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX 06 PM 34|XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX 07 PM 55|XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX 08 PM 34|XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX 09 PM 29|XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX 10 PM 33|XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX 11 PM 19|XXXXXXXXXXXXXXXXXXXXX 12 PM 49|XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX So, it looks like posting conflicts with eating, and breakfast is 6 AM, lunch at noon, supper 5-6 PM, and late snack at 11 PM.

Cream, whipping cream, and whipped cream, are some of the best foods on the planet. Whipping your own from the best whipping cream you can get, with some sugar and vanilla, is totally far over the top. If you want the whipped cream to stay whipped, then try to get some especially high quality whipping cream which may have butterfat of about 40% instead of just 36%. In New York State, I get Donnybrook, from a dairy not far away. Cheap? No. Terrific? Definitely! I whip by hand with a large stainless steel wire whip, in a stainless steel bowl with a quite rounded bottom, set in a bowl of ice water. Gee, if about to eat all those calories, then might as well work off a few whipping the cream! In Rose Levy Beranbaum, The Cake Bible, ISBN 0-688-04402-6, William Morrow, New York, 1988. is a way essentially to raise the butter fat content to make whipped cream more stable.